Knowledge Comic Creator

Create original knowledge comics with flexible art style × tone combinations.

Usage

/baoyu-comic posts/turing-story/source.md
/baoyu-comic article.md --art manga --tone warm
/baoyu-comic  # then paste content

Options

Visual Dimensions

--art

--tone

--layout

--aspect

--lang

--model

gemini

Partial Workflow Options

--storyboard-only

--prompts-only

--images-only

--regenerate N

3

2,5,8

Details: references/partial-workflows.md

Art Styles (画风)

ligne-claire

storybook-watercolor

cozy-hand-drawn-storybook-cartoon

manga

realistic

ink-brush

chalk

Tones (基调)

neutral

warm

dramatic

romantic

energetic

vintage

action

Preset Shortcuts

Presets with special rules beyond art+tone:

--style ohmsha

--art manga --tone neutral

--style wuxia

--art ink-brush --tone action

--style shoujo

--art manga --tone romantic

--style storybook

--art storybook-watercolor --tone warm

Compatibility Matrix

Details: references/auto-selection.md

Gemini Model Optimization

When

--model gemini

Key Requirements

Every prompt file MUST include:

1. Preamble block - Gemini constraints as comment header
2. Cast lock section - tokens preferred; for small casts, exact names are acceptable
3. Last check section - Validation checklist at end of each prompt

Preamble Structure

# Gemini Image Generation Best Practices (reduce distortion)
# Model: Gemini image generation (Nano Banana Pro)
# Use with: attach `{comic-dir}/characters/characters.png`
#
# TEXT QUALITY: Sans-serif, LARGE font, exact wording, leave blank if unreadable
# IDENTITY LOCK: Match character sheet exactly, preserve skin tone in lighting
# CAST LOCK: tokens preferred; exact names ok for small casts
# DIAGRAMS: Icons + short labels (≤6 words), large stat cards

Failure Mode Prevention

[CHAR_TOKEN]

Last Check Section

End every prompt with:

---
LAST CHECK BEFORE RENDERING:
- [ ] {count constraint}
- [ ] {character constraint}
- [ ] {text/visual constraint}

Full guide: references/gemini-image-prompting.md

Gemini API Setup (

--model gemini

)

For

--model gemini

GEMINI_API_KEY

GOOGLE_API_KEY

GOOGLE_API_KEY

GEMINI_API_KEY

GOOGLE_IMAGE_MODEL

gemini-3-pro-image-preview

GOOGLE_BASE_URL

Quick setup in skill env file:

mkdir -p .baoyu-skills
cat >> .baoyu-skills/.env <<'EOF'
GEMINI_API_KEY=your_key_here
# Optional:
# GOOGLE_IMAGE_MODEL=gemini-3-pro-image-preview
EOF

Auto Selection

Content signals determine default art + tone + layout (or preset):

When preset is recommended: Load

references/presets/{preset}.md

Details: references/auto-selection.md

Script Directory

Important: All scripts are located in the

scripts/

Agent Execution Instructions:

1. Determine this SKILL.md file's directory path as

   SKILL_DIR

2. Script path =

   ${SKILL_DIR}/scripts/<script-name>.ts

3. Replace all

   ${SKILL_DIR}

   in this document with the actual path

Script Reference:

scripts/merge-to-pdf.ts

scripts/gemini-image.ts

File Structure

Output directory:

comic/{topic-slug}/

• Slug: 2-4 words kebab-case from topic (e.g.,

  alan-turing-bio

  )
• Conflict: append timestamp (e.g.,

  turing-story-20260118-143052

  )

Contents:

source-{slug}.{ext}

analysis.md

storyboard.md

characters/characters.md

characters/characters.png

prompts/NN-{cover|page}-[slug].md

NN-{cover|page}-[slug].png

{topic-slug}.pdf

Language Handling

Detection Priority:

1. --lang

   flag (explicit)
2. EXTEND.md

   language

   setting
3. User's conversation language
4. Source content language

Rule: Use user's input language or saved language preference for ALL interactions:

• Storyboard outlines and scene descriptions
• Image generation prompts
• User selection options and confirmations
• Progress updates, questions, errors, summaries

Technical terms remain in English.

Workflow

Progress Checklist

Comic Progress:
- [ ] Step 1: Setup & Analyze (1.1 Preferences, 1.2 Analyze, 1.3 Check existing)
- [ ] Step 2: Confirmation - Style & options ⚠️ REQUIRED
  - [ ] 2a: Confirm style/tone/layout/reviews
  - [ ] 2b: Confirm page review rubric (multi-page projects only) → rubric.md
- [ ] Step 3: Generate storyboard + characters
- [ ] Step 4: Review outline (conditional)
- [ ] Step 5: Generate prompts
- [ ] Step 6: Review prompts (conditional)
- [ ] Step 7: Generate images ⚠️ CHARACTER REF REQUIRED
  - [ ] 7.1 Generate character sheet FIRST → characters/characters.png
  - [ ] 7.2 For each page: generate → rubric check → pass or regenerate → next
- [ ] Step 8: Merge to PDF
- [ ] Step 9: Completion report

Flow

Input → Preferences → Analyze → [Check Existing?] → [Confirm: Style + Reviews + Rubric?] → Storyboard → [Review?] → Prompts → [Review?] → Images (per-page: generate → rubric check → pass/regenerate) → PDF → Complete

Step Summary

analysis.md

rubric.md

storyboard.md

characters/

--model gemini

prompts/*.md

characters/characters.png

*.png

{slug}.pdf

Step 7: Image Generation ⚠️ CRITICAL

Character reference is MANDATORY for visual consistency.

7.1 Generate character sheet first:

# Use Reference Sheet Prompt from characters/characters.md
npx -y bun ${SKILL_DIR}/../baoyu-image-gen/scripts/main.ts \
  --promptfiles characters/characters.md \
  --image characters/characters.png --ar 4:3

If

--model gemini

npx -y bun ${SKILL_DIR}/scripts/gemini-image.ts \
  --promptfiles characters/characters.md \
  --image characters/characters.png --ar 4:3

Compress character sheet (recommended): Compress to reduce token usage when used as reference image:

• Use available image compression skill (if any)
• Or system tools:

  pngquant

  ,

  optipng

  ,

  sips

  (macOS)
• Keep PNG format, lossless compression preferred

7.2 Generate each page WITH character reference:

--ref

characters/characters.png

--ref

# Example: ALWAYS include --ref for consistency
npx -y bun ${SKILL_DIR}/../baoyu-image-gen/scripts/main.ts \
  --promptfiles prompts/01-page-xxx.md \
  --image 01-page-xxx.png --ar 3:4 \
  --ref characters/characters.png

If

--model gemini

npx -y bun ${SKILL_DIR}/scripts/gemini-image.ts \
  --promptfiles prompts/01-page-xxx.md \
  --image 01-page-xxx.png --ar 3:4 \
  --ref characters/characters.png

Full workflow details: references/workflow.md

Step 7.2: Per-Page Rubric Review

After each page is generated, run the rubric check autonomously:

1. Read rubric.md from the comic output directory
2. Review the generated image against every criterion
3. Log result clearly:
   ✅ Page N — all criteria passed → compress + proceed
   ❌ Page N — failed: [list specific failures] → adjust prompt + regenerate
4. After regeneration, re-run rubric check
5. If still failing after one regeneration → pause, show user the image + specific failures

Full rubric guide: references/page-review-rubric.md

EXTEND.md Paths

.baoyu-skills/baoyu-comic/EXTEND.md

$HOME/.baoyu-skills/baoyu-comic/EXTEND.md

EXTEND.md Supports: Watermark | Preferred art/tone/layout | Custom style definitions | Character presets | Language preference

Schema: references/config/preferences-schema.md

References

Core Templates:

• analysis-framework.md - Deep content analysis
• character-template.md - Character definition format
• storyboard-template.md - Storyboard structure
• ohmsha-guide.md - Ohmsha manga specifics
• gemini-image-prompting.md - Gemini/Nano Banana Pro optimization
• page-review-rubric.md - Per-page quality rubric template + autonomous review logic

Style Definitions:

• references/art-styles/

  - Art styles (storybook-watercolor, cozy-hand-drawn-storybook-cartoon, ligne-claire, manga, realistic, ink-brush, chalk)

• references/tones/

  - Tones (neutral, warm, dramatic, romantic, energetic, vintage, action)

• references/presets/

  - Presets with special rules (ohmsha, wuxia, shoujo)

• references/layouts/

  - Layouts (standard, cinematic, dense, splash, mixed, webtoon)

Workflow:

• workflow.md - Full workflow details
• auto-selection.md - Content signal analysis
• partial-workflows.md - Partial workflow options

Config:

• config/preferences-schema.md - EXTEND.md schema
• config/first-time-setup.md - First-time setup
• config/watermark-guide.md - Watermark configuration

Notes

• Image generation: 10-30 seconds per page
• Auto-retry once on generation failure
• Use stylized alternatives for sensitive public figures
• Maintain style consistency via session ID
• Step 2 confirmation required - do not skip
• Steps 4/6 conditional - only if user requested in Step 2
• Step 7.1 character sheet MUST be generated before pages - ensures consistency
• Step 7.2 EVERY page MUST reference characters - use

  --ref

  or embed descriptions
• Watermark/language configured once in EXTEND.md